PORT OATE February 1993 DUG. SE:LECTE»% IAY I I ns7 y(r"""«b^"

Size: px

Start display at page:

Download "PORT OATE February 1993 DUG. SE:LECTE»% IAY I I ns7 y(r"""«b^""

Isabella Hart
5 years ago
Views:

LLr in is as fnih AD-A263 699 STATION PAGE 4. TITLE AND SUBTITLE BaSQMICIIflli prrppncpc form Approved 0M8 No 0704-0188 m»t«o to i.t'jqe ' lour 0*' 'near««ncmomg <n* iim* *or wtwmg mttruaiont.

Oirtnortif for i«)orm»tion Operation»»no Ktportt. 12 IS itttmon <t 0«it*o< M»n«cjtm*«t ina lua?ft. otrwori «tauetion rojfctcotcmtu}. WnninatOA.

1 LLr in is as fnih AD-A STATION PAGE 4. TITLE AND SUBTITLE BaSQMICIIflli prrppncpc form Approved 0M8 No m»t«o to i.t'jqe ' lour 0*' 'near««ncmomg <n* iim* *or wtwmg mttruaiont. ttarcnmq * >ittii«g am «euren. d r*vi«wino int collection of rnormition i»«d comment«rmiramg tun Ouratn «limit* Of o«n«f MKI O* tm» it euro*«to «winnqion HtMau<nt't Strvicn. Oirtnortif for i«)orm»tion Operation»»no Ktportt. 12 IS itttmon <t 0«it*o< M»n«cjtm*«t ina lua?ft. otrwori «tauetion rojfctcotcmtu}. WnninatOA. OC JOWL PORT OATE February 1993 A Novel Approach for New Radiation Sources Based or Solid State Plasma Instabilities. I. AUTMOR(S) 3. REPORT TYPE AND OATES COVERED FINAL REPORT Dec.1989 to Dec S. FUNDING NUMBERS DAAL03-90-G-0015 P. Bakshi and K. Kempa 7. PERFORMING ORGANIZATION NAME(S) AND ADORESS(E Physics Department Boston College Chestnut Hill MA I. SPONSORING/MONITORING AGENCY NAME(S) AN ü. S. Army Research Office P. 0. Box Research Triangle Park, NC DUG SE:LECTE»% IAY I I ns7 y(r"""«b^" I. PERFORMING ORGANIZATION REPORT NUMBER 10. SPONSORING / MONITORING AGENCY REPORT NUMBER OfU 26*?a.A* -9H- 11. SUPPLEMENTÄRV NOTES The view, opinions and/or findings contained In this report are those of the author(s) and should not be construed as an official Department of the Army position, policy, or decision, unless so designated bv other documentation. 12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION COOE Approved for public release; distribution unlimited. IS. ABSTRACT (MiMimumiOO wortis) In this program we have developed a Novel Approach for New Radiation Sources based on current driven plasma Instabilities (CDPI) in layered solid state systems. We have demonstrated the feasibility of generating CDPI in a variety of systems, and of conversion of the energy of the growing plasma waves to electromagnetic radiation. We have identified the most promising candidates for experimental verification and device applications within the class of uniform lower dimensional systems. In addition, we have preliminary indications that modulated lower dimensional systems would significantly enhance these effects. We have also stimulated experimental efforts to confirm the results of our investigations. _._ Wpl H. SUBJECT TERMS Current n r iven Plasma Instabilities, Lower Dimension- IS. NUMBER OF PAGES 9 al Solid State Systems, Superconducting Layers, Ballistic Notion, Superlactices, Quantum Wires, Grating Couplers, Millimeter and Sub.H. PRICE CODE millimeter Waves. Device ApniieaHfln«. Ernenn of».jhh UUL 17. SECURITY CLASSIFICATION II. SECURITY CLASSIFICATION II. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF REPORT OF ABSTRACT UNCLASSIFIED UL UNCLASSIFIED UNCLASSIFIED

2 A Novel Approach for New Radiation Sources Based on Solid State Plasma Instabilities. Final Report U.S. Army Research Office Grant number DAAL03-90-G-0015 P. Bakshi and K. Kempa Physics Department Boston College Chestnut Hill MA February 1993

I. Introduction In this program we have developed a Novel Approach for New Radiation Sources based on current driven plasma instabilities (CDPI) in layered solid state systems.

We have identified the most promising candidates for experimental verification and device applications within the class of uniform lower dimensional systems.

3 I. Introduction In this program we have developed a Novel Approach for New Radiation Sources based on current driven plasma instabilities (CDPI) in layered solid state systems. We have demonstrated the feasibility of generating CDPI in a variety of systems, and of conversion of the energy of the growing plasma waves to electromagnetic radiation. We have identified the most promising candidates for experimental verification and device applications within the class of uniform lower dimensional systems. In addition, we have preliminary indications that modulated lower dimensional systems would significantly enhance these effects. We have also stimulated experimental efforts to confirm the results of our investigations. Section n provides a statement of the problem studied, section III summarizes the most important results and the last section lists all publications, and participating scientific personnel. II. Statement of the Problem The basic problem investigated in this program was a study of CDPI in solid state rystems, leading to a novel approach for generation or amplification of electromagnetic radiation in the millimeter and submillimeter ranges, with potential applications to a new class of devices. The basic principle is to utilize the energy of & dc current passing through a plasma, which leads to a plasma-instability. The instability mechanism transfers the energy from the current to characteristic plasma waves which grow in amplitude. These waves, in turn, can er radiatively decay under an appropriate set up, emitting electromagnetic radiation in a $y predictable frequency range. rj >n. In this program we chose high mobility, lower dimensional, uniform systems as the, solid state media for achieving CDPI, since they offer reduced carrier scattering effects and }/ high carrier drift velocities. -J^l^HiiT. Co<s 5!_ Avail ao.3/or Diet 1 JJ'iiLi fauiuil"i i jittjl jjutsb 1i kü Special

Our goal was to identify systems where the most efficient energy transfer from the current into the plasma waves occurs with the "mallest possible drift velocities, and where a subsequent efficient

We had investigated the feasibility of current driven instabilities in type I and type II semiconductor superlattices.

4 Our goal was to identify systems where the most efficient energy transfer from the current into the plasma waves occurs with the "mallest possible drift velocities, and where a subsequent efficient conversion of this energy into electromagnetic radiation takes place. HI. Main Results. We had investigated the feasibility of current driven instabilities in type I and type II semiconductor superlattices. The electron-phonon and electron-electron collisions were included through velocity independent collision rates. The carrier layers were taken to be strictly two dimensional (2D). For the type II superlattice (periodic arrangement of the type II heterojunctions), at liquid helium temperatures, we found that a current driven instability occurs at a drift velocity of the order of the electron Fermi velocity. The amplifiable mode is an acoustic one, which without the driving current is not observable, since it is located in the single particle absorption continuum. Frequencies corresponding to the submilimeter-wave range can be generated. Even for the liquid nitrogen temperatures this instability can be generated with slightly higher drift velocities. For the type I superlattice, we found that a new, current-induced acoustic mode comes into existence at high drifts. It can be amplified at achievable drift velocities for liquid helium temperatures. Again, frequencies corresponding to the submillimeter- wave range can be generated. Single heterojunction of type I, as compared to the superlattice arrangement, yields the same threshold drift velocity, but a lower frequency range. The physical mechanism for the instability in all these cases is the inverse Landau damping. Upon relaxing the assumption of the strictly 2D configuration of the charge carrier layers, the finite layer size creates multiple subbands. We found that, if only a single subband is occupied (in the driftless condition) in such a multisubband system, our previous results based on the strictly 2D layer model are recovered. If higher subbands are occupied (in the driftless condition), the threshold drift velocity for instability is reduced

due to the lowering of the Fermi velocity. We have also investigated the role of dimensionality by studying quantum wire systems (ID) 5. In addition, some novel systems were considered.

We found that another type of instability ( so called beam-plasma instability) occurs by passing opposite currents in alternate layers.

5 due to the lowering of the Fermi velocity. We have also investigated the role of dimensionality by studying quantum wire systems (ID) 5. In addition, some novel systems were considered. Starting from the premise that the effects of collisions can be by-passed in superconductors, we had investigated current driven instabilities in superconductor systems at liquid helium temperatures. We found that another type of instability ( so called beam-plasma instability) occurs by passing opposite currents in alternate layers. This instability does not require any threshold drift A recent experiment provides a preliminary indication of transfer of energy from the driving current into the carrier plasma in broad agreement with our theory 7. The effects of collisions can also be avoided in the ballistic motion of electrons. Streams of such electrons, with extremely high velocities, can now be experimentally achieved. We found 8 that instability is indeed possible and the physical mechanism is the inverse Landau damping. Frequencies corresponding to the submillimeter-wave range can be generated. In a somewhu different context we had noted that instabilities in the quantum resonance devices ( e.g. quantum resonant diodes) can also be viewed as current driven plasma instabilities. In May 1991 the Army Research Office hosted a "Workshop on Solid State Amplification Schemes for Electromagnetic Waves in the MMW/SMMW Region". Presentation of the results of our program and the ensuing discussions were a major part of this workshop. There was considerable interest in the gain levels of current driven plasma instabilities that could be achieved in realistic systems (we presented a preliminary account on this subject at the Workshop). Another important issue was whether the plasma waves could be efficiently coupled to electromagnetic radiation. Since then we completed our study of the growth rates of current driven plasma instabilities, and showed 10 that significant gains can be achieved in a variety of layered systems. We also developed a theory for the combined system of a grating coupler and the layered system which generates the instability, obtaining quantitatively for the systems

In addition, we have stimulated efforts towards experimental verification of these phenomena. A recent experiment here (at Boston College) based on our ideas offers a preliminary confirmation.

6 studied the efficiency for conversion of the plasma wave energy into the electromagnetic radiation 11. We also discovered that the grating can play an active role in generating a new instability, which is similar to the resistive wall instability. In addition, we have stimulated efforts towards experimental verification of these phenomena. A recent experiment here (at Boston College) based on our ideas offers a preliminary confirmation. It shows the occurrence of a drag between two layers of superconducting carriers (one of them moving). This drag could he due to a rapid dissipation of energy of carriers through spontaneous generation of plasmons. The energetic analysis of this system is in broad quantitative agreement with our theory. Efforts are also under way at Princeton University to obtain experimental verification in semiconductor-based systems 12. We have also studied 1317 novel lower dimensional systems (quantum wires and dots) for possible application as grating couplers which might offer special advantages. In contrast to the conventional metallic strip grating, a quantum wire (or dot) grating has its own characteristic frequency of oscillations, in the frequency domain of our interest. Resonant interactions will occur when this characteristic frequency matches the frequency of the current driven plasma mode in the layered system beneath the wire array, and q, the momentum of the plasmon, matches the grating period. These interactions allow for direct coupling of the system to the electromagnetic radiation. A spontaneous emission of photons will occur due to the transition back from the excited state to the ground state. In conclusion, we have now shown theoretically for various layered solid state systems that current driven plasma instabilities can be generated, plasma waves can grow and the energy can be radiated in the millimeter/submillimeter wave range. We find that the most promising semiconductor media are systems in the ballistic mode of operation 810. In a high mobility sample, this scatteringless environment is possible as long as the distance between electrodes is less than the mean free path L-lOO^m As a typical example, if the density of ballistic and stationary electrons is n,=n b =10 u cnr 2, the threshold velocity is -

$4xl0 12 s"\ which corresponds to an e-folding distance of a few hundred A.$

7 2.5xl0 7 cm/s. This threshold drift can be lowered in two ways: by lowering the density and by considering a counter-streaming ballistic arrangement, where the two layers drift in opposite directions with equal velocities. For n=2xl0 10 cnr 2, in such a counter-streaming ballistic arrangement, we only need 0.5xl0 7 cra/s as a threshold drift and even 1.0xl0 7 cm/s provides a growth rate of 1.4xl0 12 s"\ which corresponds to an e-folding distance of a few hundred A. This extremely strong instability allows an amplification of several orders of magnitude, even for a lojim range of ballistic operation. For a typical value 10 of the plasma wave number 4x10 s cm- 1, the grating period to quench the momentum is about 1600Ä. The corresponding frequency is greater than 2xl0 12 s 1, and the wavelength of the emitted radiation is in the millimeter to submillimeter regime. Typical amplitude conversion efficiency of the standard metallic strip grating 11 is about 1%. Additional feedback mechanisms through a surrounding resonant cavity can further enhance the efficiency. The threshold drifts in the non-ballistic mode of operation are higher and, while still achievable within the current state of the art technology, pose a practical problem. Our cumulative experience on this program leads us to suggest that ultimately the best systems could be the modulated lower dimensional systems. They can offer a twofold advantage: the threshold drift velocities are significantly lowered, and the modulation acts as a built-in grating. Our recently initiated preliminary studies confirm this. References for Section III 1. P. Bakshi, J. Cen and K. Kempa," Amplification of Surface Modes in Type II Semiconductor Superlattices.", J. Appl. Phys.64,2243 (1988). 2. J. Cen, K. Kempa and P. Bakshi," Amplification of a New Surface Mode in Type I Semiconductor Superlattices.", Phys. Rev. B 38,10051 (1988). 3. K. Kempa, P. Bakshi and J. Cen, "Plasmon Amplification in Semiconductor Superlattices" Proceedings of SPIE, vol. 945,62, (1988).

4. J. Cen, K. Kempa, and P. Bakshi, "Amplification of Plasma Modes in Semiconductor Heterostructures", Solid State Comm. 78,433-437, (1991). 5. P. Bakshi, J. Cen and K.

Bakshi, " Current-Driven Plasma Instabilities in Superconductors", Phys. Rev.B39, 2852, (1989). 7. F. Dong, M.

8 4. J. Cen, K. Kempa, and P. Bakshi, "Amplification of Plasma Modes in Semiconductor Heterostructures", Solid State Comm. 78, , (1991). 5. P. Bakshi, J. Cen and K. Kempa," Current driven plasma instability in quantum wires", Solid State Communications 76, , (1990). 6. K. Kempa, J. Cen, P. Bakshi, " Current-Driven Plasma Instabilities in Superconductors", Phys. Rev.B39, 2852, (1989). 7. F. Dong, M.J. Graf, and P.M. Mankiewich, "Experimental search for current driven plasma instabilities in superconducting layers", Solid State Comm. 84, , (1992) 8. K. Kempa, P. Bakshi, J. Cen and H. Xie, "Spontaneous Generation Of Plasmons by Ballistic Electrons", Phys. Rev. B43, , (1991). 9. P. Bakshi and K. Kempa," Possible new mechanism for plasma instabilities in solid state devices", Phys. Rev. B 40,3433, (1989). 10. H. Xie, K. Kempa and P. Bakshi," Growth rates of current excited plasma waves in semiconductor layered systems", J. Appl. Phys. 72,4767, (1992) 11. K. Kempa, P. Bakshi, H. Xie, and W.L. Schaich, " Current driven plasma instabilities in solid state layered systems with a grating", Phys.Rev.B 47, 4532, (1993). 12. D. Tsui and K. Hirakawa, private communication. 13. D. A. Broido, P. Bakshi and K. Kempa. "Electromagnetic Response of Lower dimensional electron systems: Implications of omitting the ground state calculations", Solid State Comm. 76, , (1990). 14. P. Bakshi, D.A. Broido and K. Kempa," Electromagnetic response of quantum dots", Phys. Rev. B42, 7416, (1990). 15. D.A. Broido, K. Kempa, and P. Bakshi," Self-consistent far infrared response of quantum dot structures", Phys. Rev. B 42,11400, (1990).

9 16. K. Kempa, D.A. Broido, and P. Bakshi," Spontaneous polarization in quantum dot systems", Phys. Rev. B 43,9343, (1991). 17. P. Bakshi, D.A. Broido and K. Kempa," Spontaneous polarization of electrons in quantum dashes", J. Appl. Phys. 70,5150, (1991).

IV. Publications and Personnel. a) Publications 1. P. Bakshi, J. Cen and K. Kempa," Current driven plasma instability in quantum wires", Solid State Communications 76,835-837, (1990). 2. P. Bakshi, D.

Solid State Comm. 76,613-616, (1990). 4. D.A. Broido, K. Kempa, and P Bakshi," Self-consistent far infrared response of quantum dot structures", Phys. Rev. B 42,11400, (1990). 5. J. Cen, K.

10 IV. Publications and Personnel. a) Publications 1. P. Bakshi, J. Cen and K. Kempa," Current driven plasma instability in quantum wires", Solid State Communications 76, , (1990). 2. P. Bakshi, D.A. Broido and K. Kempa," Electromagnetic response of quantum dots", Phys. Rev. B42, 7416, (1990). 3. D. A. Broido, P. Bakshi and K. Kempa. "Electromagnetic Response of Lower dimensional electron systems: Implications of omitting the ground state calculations". Solid State Comm. 76, , (1990). 4. D.A. Broido, K. Kempa, and P Bakshi," Self-consistent far infrared response of quantum dot structures", Phys. Rev. B 42,11400, (1990). 5. J. Cen, K. Kempa, and P. Bakshi, "Amplification if Plasma Modes in Semiconductor Heterostructures", Solid State Comm. 78, , (1991). 6. K. Kempa, P. Bakshi, J. Cen and H. Xie, "Spontaneous Generation Of Plasmons by Ballistic Electrons", Phys. Rev. B43, , (1991). 7. K. Kempa, D.A. Broido, and P. Bakshi," Spontaneous polarization in quantum dot systems", Phys. Rev. B 43,9343. (1991). 8. P. Bakshi, D.A. Broido and K. Kempa," Spontaneous polarization of electrons in quantum dashes", J. Appl. Phys (1991). 9. H. Xie, K. Kempa and P. Bakshi," Growth rates of current excited plasma waves in semiconductor layered systems", J. Appl. Phys. 72,4767, (1992). 10. K. Kempa, P. Bakshi, H. Xie, and W.L. Schaich, " Current driven plasma instabilities in solid state layered systems with a grating", Phys.Rev.B 47,4532, (1993).

Xie (Research Assistant) c) Degree earned J. Cen, Ph.

11 b) Personnel 1. P. Bakshi (Faculty), Principal Investigator 2. K. Kerapa (Faculty), Principal Investigator 3. J. Cen (Research Assistant) 4. H. Xie (Research Assistant) c) Degree earned J. Cen, Ph.D M Boston College, 1991 Thesis title: "Current driven plasma instabilities in layered solid state systems" d) Reportable inventions. None

Defense Technical Information Center Compilation Part Notice

UNCLASSIFIED Defense Technical Information Center Compilation Part Notice ADP012759 TITLE: Low-Dimensional Electron Semiconductor Structures as Tunable Far-Infrared Amplifiers and Generators DISTRIBUTION: